An epic voyage in the making

Nature Geoscience 2, 733 (2009). doi:10.1038/ngeo669

The plan to drill through the entire oceanic crust is ambitious and exciting, and well worth the expense.

Nature Geoscience 2, 733 (2009). doi:10.1038/ngeo688

As climate change continues to erode biodiversity, the two disciplines need to improve their dialogue.

Nature Geoscience 2, 737 (2009). doi:10.1038/ngeo671

Authors: G. R. van der Werf, D. C. Morton, R. S. DeFries, J. G. J. Olivier, P. S. Kasibhatla, R. B. Jackson, G. J. Collatz & J. T. Randerson

Deforestation is the second largest anthropogenic source of carbon dioxide to the atmosphere, after fossil fuel combustion. Following a budget reanalysis, the contribution from deforestation is revised downwards, but tropical peatlands emerge as a notable carbon dioxide source.

Nature Geoscience 2, 741 (2009). doi:10.1038/ngeo673

Author: Bruce Buffett

The use of more realistic parameters in numerical geodynamo simulations tends to generate less Earth-like magnetic fields. This paradox could be resolved by considering uniform heat flux instead of uniform temperature at the core's surface.

Nature Geoscience 2, 742 (2009). doi:10.1038/ngeo678

Author: Martyn Chipperfield

The stratospheric ozone layer has undergone severe depletion as a result of anthropogenic halocarbons. Although the Montreal Protocol has provided relief, anthropogenic emissions of another substance, nitrous oxide, are set to dominate ozone destruction.

Nature Geoscience 2, 743 (2009). doi:10.1038/ngeo676

Author: Niels Nørgaard-Pedersen

Sea ice is an integral component of the climate system, but a difficult one to reconstruct. Biochemical tracers preserved in marine sediments now reveal the waxing and waning of sea ice since the Last Glacial Maximum in an Arctic Ocean gateway.

Nature Geoscience 2, 744 (2009). doi:10.1038/ngeo674

Author: Magali I. Billen

Faults that develop in subducting slabs act as conduits for sea water. Numerical modelling indicates that pressure gradients resulting from the bending of slabs may then drive the water deep into their interior.

Nature Geoscience 2, 747 (2009). doi:10.1038/ngeo675

Author: Mikaël Attal

Mountain landscapes are shaped by tectonics and climate. A series of laboratory experiments has documented a mechanism by which mountain river networks split as the geometry of a mountain evolves in response to an orographic precipitation gradient.

Nature Geoscience 2, 749 (2009). doi:10.1038/ngeo677

Author: Elisa Manzini

The El Niño/Southern Oscillation phenomenon is the most prominent source of climate variability. Emerging evidence suggests that its signature is not limited to the lower layers of the atmosphere.

Nature Geoscience 2, 751 (2009). doi:10.1038/ngeo660

Authors: P. C. Tzedakis, D. Raynaud, J. F. McManus, A. Berger, V. Brovkin & T. Kiefer

Nature Geoscience 2, 757 (2009). doi:10.1038/ngeo662

Authors: Alejandro Luque & Ute Ebert

Sprite discharges above thunderclouds at altitudes of 40–90 km (refs 1, 2, 3, 4, 5) are usually created by a strong positive cloud-to-ground lightning flash. Sometimes these sprite discharges emerge from a visible halo, and during the first stage they always propagate downwards and branch on their way. Modelling efforts have been restricted to conditions of non-ionized air of constant density and show double-headed sprites or sprites starting from metal electrodes, but they do not explain why observations exclusively record sprites that propagate downwards. Here we present simulations with a numerical discharge model on a non-uniform, dynamically adapted computational grid to capture the wide range of emerging spatial scales, and we use realistic air and electron densities that vary with altitude. Our model shows a downward-propagating screening-ionization wave in the lower ionosphere that sharpens and collapses into a sprite streamer as it propagates farther down. Streamer velocity, diameter and length until branching agree with observations within measuring accuracy. We speculate that sprites generically emerge through the collapse of a wide screening-ionization wave into a sprite streamer, although this wave is only sometimes visible as a luminous halo.

Nature Geoscience 2, 761 (2009). doi:10.1038/ngeo657

Authors: J. N. Moum, R.-C. Lien, A. Perlin, J. D. Nash, M. C. Gregg & P. J. Wiles

Changes in sea surface temperature of equatorial waters have critical effects on the large-scale atmospheric circulation. So far, large-scale, energetic tropical instability waves in equatorial waters have been thought to warm the sea surface through both meridional and zonal advection. Here, we present shipboard profiling measurements of turbulence kinetic-energy dissipation rate that reveal unanticipated vigorous mixing associated with tropical instability waves. The meridional tropical instability-wave shear increases the shear above the core of the Equatorial Undercurrent, which is already large, nudging the flow toward instability. As a consequence, turbulence dissipation rates and heat fluxes are many times greater than previous measurements at the same location but in the absence of tropical instability waves. The vertical divergence of turbulence heat flux is sufficient to cool the upper layer by 2 K per month, and heat the core of the Equatorial Undercurrent by 10 K per month. Long-term records at 140∘ W further reveal that cooling of the sea surface is significantly correlated to tropical-instability-wave kinetic energy. Thus, seasonal surface cooling in the central equatorial Pacific may be largely caused by mixing induced by tropical instability waves.

Nature Geoscience 2, 766 (2009). doi:10.1038/ngeo666

Author: Stephane Bonnet

Climate, and in particular the spatial pattern of precipitation, is thought to affect the topographic and tectonic evolution of mountain belts through erosion. Numerical model simulations have suggested that the main drainage divide continuously migrates and asymmetric topography in mountain ranges develops in response to horizontal tectonic motion or orographic precipitation. The effects of such a migration have, however, been challenging to observe in natural settings. Here I document the effects of a lateral precipitation gradient on a landscape undergoing constant uplift in a laboratory modelling experiment. In the experiment, the drainage divide migrates towards the drier, leeward side of the mountain range, causing the drainage basins on the leeward side to shrink and split into smaller basins. This mechanism results in a progressively increasing number of drainage basins on the leeward side of the mountain range as the divide migrates, such that the expected relationship between the spacing of drainage basins and the location of the main drainage divide is maintained. I propose that this mechanism could clarify the drainage-divide migration and topographic asymmetry found in active orogenic mountain ranges, as seen in the Aconquija Range of Argentina.

Nature Geoscience 2, 772 (2009). doi:10.1038/ngeo665

Authors: Juliane Müller, Guillaume Massé, Rüdiger Stein & Simon T. Belt

Sea ice is a critical component of the climate system: variations in sea-ice cover affect the albedo of polar regions, and also the rate of deepwater formation. Changes in the sea-ice cover of the North Atlantic Ocean are thought to have been related to abrupt climate changes throughout the last glacial termination, but reconstructions of sea-ice conditions are rare. Here we use the sedimentary abundance of the IP25 and brassicasterol biomarkers, produced by sea-ice-associated diatoms and open-water phytoplankton, respectively, to generate a record of sea-ice conditions in the northernmost Atlantic Ocean for the past 30,000 years. Our reconstruction shows that a stationary margin between sea-ice cover and the open ocean existed during the Last Glacial, although perennial sea-ice cover prevailed for most of the Last Glacial Maximum. An early warming about 14,000 years ago was associated with ice-free conditions; however, seasonal sea ice was present throughout the Holocene. We find temporal links between our record of sea ice and reconstructions of the amount of relatively warm Atlantic water advected into the Nordic Seas. We therefore conclude that changes in sea-ice conditions are linked to regional and global climate anomalies and oceanographic circulation in the North Atlantic.

Nature Geoscience 2, 777 (2009). doi:10.1038/ngeo668

Authors: Appy Sluijs, Stefan Schouten, Timme H. Donders, Petra L. Schoon, Ursula Röhl, Gert-Jan Reichart, Francesca Sangiorgi, Jung-Hyun Kim, Jaap S. Sinninghe Damsté & Henk Brinkhuis

Several episodes of abrupt and transient warming, each lasting between 50,000 and 200,000 years, punctuated the long-term warming during the Late Palaeocene and Early Eocene (58 to 51 Myr ago) epochs. These hyperthermal events, such as the Eocene Thermal Maximum 2 (EMT2) that took place about 53.5 Myr ago, are associated with rapid increases in atmospheric CO2 content. However, the impacts of most events are documented only locally. Here we show, on the basis of estimates from the TEX86′ proxy, that sea surface temperatures rose by 3–5 ∘C in the Arctic Ocean during the EMT2. Dinoflagellate fossils demonstrate a concomitant freshening and eutrophication of surface waters, which resulted in euxinia in the photic zone. The presence of palm pollen implies that coldest month mean temperatures over the Arctic land masses were no less than 8 ∘C, in contradiction of model simulations that suggest hyperthermal winter temperatures were below freezing. In light of our reconstructed temperature and hydrologic trends, we conclude that the temperature and hydrographic responses to abruptly increased atmospheric CO2 concentrations were similar for the ETM2 and the better-described Palaeocene–Eocene Thermal Maximum, 55.5 Myr ago.

Nature Geoscience 2, 781 (2009). doi:10.1038/ngeo652

Authors: Yifeng Wang, Huifang Xu, Enrique Merino & Hiromi Konishi

The chemical signatures and mineralogy of banded iron formations have the potential to provide information about the ocean environment on early Earth. Their formation requires iron- and silicon-rich fluids, but the mechanisms by which the alternating layers of Si- and Fe-rich rock formed remain controversial. Here we use thermodynamic calculations to show that Fe- and Si-rich fluids can be generated by hydrothermal leaching of low-Al oceanic crustal rocks such as komatiites. We find that positive feedbacks occur among the chemical reactions when hydrothermal fluids mix with ambient sea water. These feedbacks lead to alternating precipitation of Fe and Si minerals, owing to the formation of complexes between Fe(II) and silicic acid. We suggest that the small-scale (<1 cm) banding was produced by internal dynamics of the geochemical system, rather than any external forcing. As the Archaean eon progressed, the oceanic crust produced was rich in Al. When Al-rich crust undergoes hydrothermal alteration, Fe is locked in Al–Fe silicate minerals. This results in iron-depleted hydrothermal fluids, and thus prevents the deposition of Fe-rich minerals. We therefore conclude that the widespread cessation of banded iron formation deposition 1.7 billion years ago reflects the changing composition of the oceanic crust.

Nature Geoscience 2, 785 (2009). doi:10.1038/ngeo661

Authors: Graham J. Hill, T. Grant Caldwell, Wiebke Heise, Darren G. Chertkoff, Hugh M. Bibby, Matt K. Burgess, James P. Cull & Ray A. F. Cas

Three prominent volcanoes that form part of the Cascade mountain range in Washington State (USA)—Mounts St Helens, Adams and Rainier—are located on the margins of a mid-crustal zone of high electrical conductivity. Interconnected melt can increase the bulk conductivity of the region containing the melt, which leads us to propose that the anomalous conductivity in this region is due to partial melt associated with the volcanism. Here we test this hypothesis by using magnetotelluric data recorded at a network of 85 locations in the area of the high-conductivity anomaly. Our data reveal that a localized zone of high conductivity beneath this volcano extends downwards to join the mid-crustal conductor. As our measurements were made during the recent period of lava extrusion at Mount St Helens, we infer that the conductivity anomaly associated with the localized zone, and by extension with the mid-crustal conductor, is caused by the presence of partial melt. Our interpretation is consistent with the crustal origin of silicic magmas erupting from Mount St Helens, and explains the distribution of seismicity observed at the time of the catastrophic eruption in 1980 (refs 9, 10).

Nature Geoscience 2, 790 (2009). doi:10.1038/ngeo656

Authors: Manuele Faccenda, Taras V. Gerya & Luigi Burlini

Bending of oceanic plates at subduction zones results in extension and widespread normal faulting in the upper, brittle part of the slab. Detailed seismic surveys at trenches reveal that this part of the oceanic plate could be pervasively hydrated for several kilometres below the crust–mantle boundary. Similarly, heat-flow surveys indicate active fluid circulation within the slab. Yet, the mechanisms that enable fluids to percolate to such depths in spite of their natural buoyancy remain unclear. Here we use two-dimensional numerical experiments to show that stress changes induced by the bending oceanic plate produce subhydrostatic or even negative pressure gradients along normal faults, favouring downward pumping of fluids. The fluids then react with the crust and mantle surrounding the faults and are stored in the form of hydrous minerals. We suggest that this process is the dominant mechanism of deep slab hydration, although it may be locally aided by the enhancement in porosity due to prefailure dilatancy, pre-existing cracks and migrating fluid-filled cracks. Our results have implications for the transport of water into the deeper parts of the mantle, and for further clarifying the seismic anisotropy of slabs.

Nature Geoscience 2, 794 (2009). doi:10.1038/ngeo663

Authors: Simon A. Hunt, Donald J. Weidner, Li Li, Liping Wang, Nicolas P. Walte, John P. Brodholt & David P. Dobson

The lowermost part of the Earth’s mantle—the ∼200-km-thick D′′ layer—shows anomalous seismic properties, and is rheologically distinct from the rest of the lower mantle. The difference is thought to result from a phase transition from silicate perovskite to silicate post-perovskite. However, the rheology of the latter phase remains to be documented owing to experimental difficulties in reproducing pressures equivalent to those in the lowermost mantle. Here we address this problem by conducting laboratory experiments that use calcium iridate, which has been shown to be an appropriate low-pressure analogue. We find that the post-perovskite phase of this analogue is approximately five times weaker than its perovskite phase, and that it further weakens by a factor of two during the phase transformation; these are minimum estimates. If, as is likely, a similar weakening occurs in lower-mantle magnesium–silicate compositions, this could provide an explanation for the behaviour of the lowermost mantle as inferred from geophysical data.

Nature Geoscience 2, 798 (2009). doi:10.1038/ngeo658

Authors: James M. Brenan & William F. McDonough

The abundances of the highly siderophile elements as well as their relative proportions in the mantle deviate from those predicted by equilibrium partitioning between metal and silicate during the formation of the Earth’s core. This discrepancy is generally explained by invoking the addition of a late veneer of extraterrestrial material to the mantle after core formation was complete. Recently reported partition coefficients for gold, platinum and palladium could result in mantle abundances consistent with equilibrium partitioning. However, whether these results can be extrapolated to all highly siderophile elements, and thereby preclude the need for a late veneer, remains to be verified. Here we use high-temperature experiments to determine the metal–silicate partition coefficients for osmium, iridium and gold. On the basis of our estimates, equilibrium partitioning during core formation can explain the observed concentration of gold in the mantle, but not that of osmium and iridium. We conclude that not all highly siderophile elements were affected by core formation in the same way, and that the abundances of elements such as osmium and iridium require the addition of a late veneer.

Nature Geoscience 2, 802 (2009). doi:10.1038/ngeo643

Authors: Ataru Sakuraba & Paul H. Roberts

The Earth’s main magnetic field is thought to be generated by motions in the planet’s fluid outer core, which lead to an effect similar to that of a dynamo. Recent high-resolution numerical simulations produce only a non-dipolar or a dipolar but comparatively weak magnetic field unlike that of the Earth. Older models that did generate a strong, Earth-like field needed to use unrealistically high viscosities for the core fluid. Common to most of the models is the assumption of a laterally uniform core-surface temperature. Here we use a low-viscosity geodynamo model to evaluate the effect of a different and more realistic boundary condition—a uniform heat flux at the surface of the core—on the simulation of an Earth-like magnetic field. Our results show that when the surface temperature is laterally uniform, only a weak magnetic field is generated because planetary-scale fluid circulations are suppressed. In contrast, a laterally uniform heat flux at the core’s surface leads to large-scale convective flows, and a comparatively strong dipole-type magnetic field. Contrary to previous work, we suggest that thermal conditions at the core surface have a strong effect on low-viscosity geodynamo models.